INDEX

Explanations

occurrences of the word "supporters."

oai_token-act-pair · gpt-3.5-turbo

mentions of supporters in various contexts

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

 geological

-0.64

smith

-0.63

 Mamm

-0.62

 Surgery

-0.62

ered

-0.62

 Injury

-0.62

pher

-0.60

PASS

-0.59

ospital

-0.58

 situational

-0.58

POSITIVE LOGITS

hesis

1.05

hetical

0.97

Supporters

0.91

 supporters

0.90

 Supporters

0.82

heses

0.81

yip

0.81

 supporter

0.81

hip

0.78

hips

0.78

Activations Density 0.028%

occurrences of the word "supporters."

mentions of supporters in various contexts

No Comments

No Known Activations