INDEX

Explanations

words related to responsibility and responsive actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res_post_32k-oai

Configuration

jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_6.pt

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

32,768

Data Type

torch.float32

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fare

-0.83

 Franco

-0.66

 Bloom

-0.65

 Dresden

-0.61

 Petraeus

-0.61

nz

-0.60

 Murdoch

-0.60

Jet

-0.59

 Blast

-0.59

 Reloaded

-0.59

POSITIVE LOGITS

ibilities

1.39

Resp

1.16

ibly

1.13

ible

1.07

onding

1.07

ibility

1.02

essor

1.01

idy

0.97

umption

0.96

umed

0.93

Activations Density 0.213%

words related to responsibility and responsive actions

No Comments

No Known Activations