INDEX

Explanations

occurrences of the word "ban" followed by a number indicating the strength of the match

oai_token-act-pair · gpt-3.5-turbo

discussions about bans or prohibitions 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Generations

-0.81

LV

-0.70

ORY

-0.69

rious

-0.68

Io

-0.66

Sea

-0.66

ACTED

-0.65

 Barg

-0.65

everal

-0.63

 Fault

-0.62

POSITIVE LOGITS

hammer

1.15

zai

1.07

ishment

1.06

quet

0.91

 bans

0.89

ish

0.86

ban

0.84

jo

0.82

 banning

0.82

etooth

0.81

Activations Density 0.013%

occurrences of the word "ban" followed by a number indicating the strength of the match

discussions about bans or prohibitions in various contexts

No Comments

No Known Activations