INDEX

Explanations

phrases indicating an intention or effort towards a specific action

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "attempt" 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 @ 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

today

-0.73

 Wrap

-0.72

 Solitaire

-0.70

houses

-0.67

eyes

-0.65

minus

-0.65

effect

-0.63

saw

-0.60

riger

-0.59

front

-0.59

POSITIVE LOGITS

Attempts

1.01

 unsuccessfully

0.97

 attempt

0.93

 attempts

0.92

ossibility

0.80

ossible

0.75

orneys

0.75

 attempted

0.74

 futile

0.71

 succeed

0.69

Activations Density 0.031%

phrases indicating an intention or effort towards a specific action

occurrences of the word "attempt" in various contexts

No Comments

No Known Activations