INDEX

Explanations

phrases related to identification of individuals in different situations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "identified" and other terms related to the identification of individuals

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 enjoyment

-0.67

Interstitial

-0.64

yrim

-0.64

andel

-0.64

istas

-0.64

 patronage

-0.63

vable

-0.61

hibition

-0.61

 experimentation

-0.61

 negativity

-0.60

POSITIVE LOGITS

 surn

0.90

 redacted

0.87

 deceased

0.84

IFIED

0.81

 names

0.79

 pseudonym

0.78

Names

0.75

 Identified

0.75

Rus

0.74

©¶æ¥µ

0.74

Activations Density 0.088%

phrases related to identification of individuals in different situations

instances of the word "identified" and other terms related to the identification of individuals

No Comments

No Known Activations