INDEX

Explanations

words related to disturbing or unsettling situations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "disturbing" and its context within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

cellence

-0.79

oned

-0.76

gs

-0.73

oled

-0.72

arta

-0.72

eva

-0.72

ophers

-0.71

tes

-0.69

ht

-0.68

bid

-0.68

POSITIVE LOGITS

 disturbing

0.91

 disturb

0.81

 sexist

0.75

 undermin

0.73

urnal

0.71

 creeps

0.69

 indecent

0.69

 pornographic

0.67

 unsettling

0.67

 alarming

0.66

Activations Density 0.011%

words related to disturbing or unsettling situations

instances of the word "disturbing" and its context within the text

No Comments

No Known Activations