INDEX

Explanations

strong emotional words related to verbal interactions

oai_token-act-pair · gpt-3.5-turbo

instances of verbal interactions or communications

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

arent

-0.85

akeru

-0.84

enthal

-0.83

ress

-0.81

roxy

-0.80

ocrates

-0.78

resses

-0.76

alach

-0.73

annis

-0.72

 Profit

-0.71

POSITIVE LOGITS

 altercation

0.98

ized

0.94

ãĥ£

0.93

isations

0.91

ization

0.87

 cues

0.86

izational

0.85

 verbal

0.85

izations

0.84

izing

0.81

Activations Density 0.014%

strong emotional words related to verbal interactions

instances of verbal interactions or communications

No Comments

No Known Activations