INDEX

Explanations

the word "call" followed by a high positive activation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "call."

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

 istg

-0.84

bilt

-0.81

 embr

-0.70

inth

-0.69

bourne

-0.67

olitics

-0.66

ynski

-0.64

cffff

-0.61

ipeg

-0.61

inh

-0.60

POSITIVE LOGITS

backs

1.01

igraph

0.98

call

0.95

phas

0.86

 bullshit

0.83

calling

0.81

 Calling

0.80

 bluff

0.78

oused

0.76

Activations Density 0.053%

the word "call" followed by a high positive activation

instances of the word "call."

No Comments

No Known Activations