INDEX

Explanations

words related to failures or negative outcomes

oai_token-act-pair · gpt-3.5-turbo

references to failures 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

rete

-0.69

enfranch

-0.68

bern

-0.68

selves

-0.67

population

-0.66

atu

-0.66

riel

-0.65

utra

-0.65

rosse

-0.65

irin

-0.64

POSITIVE LOGITS

 miser

1.30

 dism

0.82

DEV

0.81

 failures

0.79

 catast

0.78

Failure

0.78

 horribly

0.77

afe

0.73

lust

0.72

fail

0.71

Activations Density 0.029%

words related to failures or negative outcomes

references to failures in various contexts

No Comments

No Known Activations