INDEX

Explanations

words related to specific targets or goals

oai_token-act-pair · gpt-3.5-turbo

instances of the word "target" and its related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UGE

-0.68

fo

-0.68

 Geological

-0.67

IGH

-0.67

ISTORY

-0.67

 notor

-0.62

ERROR

-0.62

plet

-0.61

ansk

-0.61

AX

-0.60

POSITIVE LOGITS

ted

1.13

 targets

0.97

 target

0.92

nels

0.81

izen

0.80

oided

0.77

eers

0.76

ishi

0.75

target

0.73

ataka

0.72

Activations Density 0.016%

words related to specific targets or goals

instances of the word "target" and its related contexts

No Comments

No Known Activations