INDEX

Explanations

words related to holding law enforcement accountable for their actions

oai_token-act-pair · gpt-3.5-turbo

conjunctions and modifiers indicating conditions or relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Cheong

-0.79

amination

-0.72

hov

-0.71

igl

-0.71

lang

-0.70

ificial

-0.69

ipel

-0.67

mson

-0.65

ulhu

-0.61

Day

-0.60

POSITIVE LOGITS

 obligation

0.78

 barg

0.75

 wanting

0.75

 longing

0.73

 aspiration

0.72

 fulfilled

0.71

 justification

0.71

 expectation

0.71

 compulsion

0.71

 justifies

0.70

Activations Density 0.458%

words related to holding law enforcement accountable for their actions

conjunctions and modifiers indicating conditions or relationships

No Comments

No Known Activations