INDEX

Explanations

verbs related to actions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "doing" 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lights

-0.69

sg

-0.66

wit

-0.64

antha

-0.62

pu

-0.61

untu

-0.61

ules

-0.61

liner

-0.61

dt

-0.61

coe

-0.59

POSITIVE LOGITS

berman

0.76

 nothing

0.74

 brisk

0.73

å¥³

0.73

 something

0.70

 terribly

0.70

THING

0.70

 things

0.68

LE

0.67

 alright

0.67

Activations Density 0.050%

verbs related to actions

occurrences of the phrase "doing" in various contexts

No Comments

No Known Activations