INDEX

Explanations

terms related to a negative trend or decrease in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "decline" and related concepts of reduction or decrease

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ropri

-0.68

":[{"

-0.67

WI

-0.67

RA

-0.67

bley

-0.66

racuse

-0.66

aptic

-0.65

Bet

-0.64

onen

-0.64

Pur

-0.63

POSITIVE LOGITS

cember

0.78

rates

0.77

 precip

0.75

 uranium

0.74

 escal

0.72

 attrition

0.71

 spiral

0.71

lust

0.70

 rates

0.69

 Rates

0.67

Activations Density 0.021%

terms related to a negative trend or decrease in various contexts

instances of the word "decline" and related concepts of reduction or decrease

No Comments

No Known Activations