INDEX

Explanations

mentions of fear or intimidation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "scare" and its variations

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

•layer 5 - fear

No Comments

Negative Logits

cially

-0.76

odore

-0.71

mine

-0.69

 refin

-0.67

iquette

-0.67

 offic

-0.66

avis

-0.66

ophers

-0.66

 swaps

-0.66

 slightest

-0.64

POSITIVE LOGITS

crow

1.73

mong

1.29

ingly

0.86

warts

0.80

Cry

0.77

escape

0.73

bite

0.72

 Beir

0.71

inventoryQuantity

0.69

Mew

0.69

Activations Density 0.020%

mentions of fear or intimidation

instances of the word "scare" and its variations

No Comments

No Known Activations