INDEX

Explanations

adverbs that intensify an accompanying adjective or verb

oai_token-act-pair · gpt-3.5-turbo

instances of the word "incredibly" and its variations used to express strong feelings or reactions

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

icip

-0.75

icipated

-0.73

Runner

-0.71

eto

-0.69

 Continued

-0.67

igi

-0.66

oral

-0.65

osate

-0.65

edom

-0.64

igate

-0.63

POSITIVE LOGITS

 expensive

0.77

 uncomfortable

0.76

 important

0.75

 difficult

0.75

 fascinating

0.73

 complicated

0.72

 intricate

0.72

 valuable

0.72

beit

0.72

 inefficient

0.71

Activations Density 0.026%

adverbs that intensify an accompanying adjective or verb

instances of the word "incredibly" and its variations used to express strong feelings or reactions

No Comments

No Known Activations